On 16 October 2026, Hogan Lovells Cadwalader became one of the largest law firms to publicly commit to a firmwide AI agent strategy. The firm announced it had adopted Harvey across its Corporate and Finance, Litigation and Arbitration, and Regulatory and IP practices, alongside a parallel agreement with Legora. More significantly, its own lawyers have built custom AI agents for Knowledge Management, Capital Markets, and International Arbitration, complete with collaborative playbooks.
This is not a pilot. It is a structural bet on AI agents as the operating model for legal work, backed by new senior leadership, an AI board, and a dedicated commercial unit. For enterprise buyers watching where agentic AI is landing in production, the Hogan Lovells Cadwalader deployment is a signal that professional services are now designing workflows around agents, not simply adding them to existing ones.
What the firm has done
The firm's approach has three layers that are worth unpacking.
First, the Harvey deployment is firmwide and practice-group specific. Teams have embedded the platform through practice-group office hours, tailored workshops, and individual sessions with senior lawyers. The aim has been to make the tool part of everyday legal work rather than a feature that sits in the background.
Second, the firm is simultaneously rolling out Legora, making it one of the few major law firms to publicly adopt two AI agent platforms at once. Dual-vendor strategies are rare in legal services because of integration overhead and data governance complexity. The decision suggests Hogan Lovells Cadwalader is treating AI agents as a core infrastructure layer worth diversifying, rather than a point solution to be consolidated.
Third, and most telling, the firm has moved from consuming off-the-shelf AI products to building its own agents. Lawyers have developed custom agents for Knowledge Management, Capital Markets, and International Arbitration. They have also built collaborative playbooks, which means the firm is encoding its own intellectual property into reusable agentic workflows. This is the transition from using AI to augment expertise, to using agents as a delivery mechanism for that expertise.
Why it chose two platforms
The dual-vendor decision is the most unusual part of the announcement. Most large firms pick one AI platform and build around it. Hogan Lovells Cadwalader's choice to run both Harvey and Legora reflects a growing view among sophisticated buyers that the agentic AI market is still too early to bet on a single stack.
Harvey is the established name in legal AI, with deep integrations into legal research, document drafting, and case analysis. Legora brings a different architecture focused on agentic workflow automation across legal tasks. By adopting both, the firm is hedging against platform lock-in and gaining exposure to two different approaches to the same problem: how to turn legal expertise into scalable, governable agentic output.
For enterprise buyers outside legal services, the lesson is that vendor selection in agentic AI is becoming more complex, not simpler. The right stack may not be one platform. It may be a carefully governed federation of agents running on different infrastructure, each chosen for the specific type of work it handles best.
Custom agents as intellectual property
The custom agents are the part of this story that generalises beyond law. Hogan Lovells Cadwalader has recognised that its competitive advantage is not just the knowledge individual lawyers hold, but how that knowledge is captured, structured, and applied at scale.
A custom agent for International Arbitration, for example, does not just summarise documents. It encodes the firm's accumulated pattern recognition: how tribunals have ruled on specific clauses, which arguments have succeeded in particular jurisdictions, how opposing counsel has behaved in past cases. That is proprietary data, and by building it into an agent, the firm is turning experience into infrastructure.
Miguel Zaldivar, CEO of Hogan Lovells Cadwalader, made the point directly: "Through Harvey, we're perfecting the ways to capture and apply the expertise that exists across our firm on a variety of legal work. That has helped our lawyers work more effectively and strengthens the support and advice we provide to clients."
John Haddock, Chief Business Officer at Harvey, added: "Hogan Lovells Cadwalader has a clear vision for how Harvey can elevate its lawyers' work and client service. We're proud to build alongside a firm with that ambition and help shape what comes next."
The governance architecture
The firm did not just buy software. It created an AI board, appointed new senior leadership, and formed a commercial unit to oversee the programme. That governance layer matters because it signals that the firm understands agentic AI is a business transformation, not an IT initiative.
Enterprise buyers should pay attention to this. The firms that are succeeding with agentic AI are the ones that have assigned clear accountability for outcomes, built cross-functional oversight, and treated the technology as a strategic investment rather than a cost centre. Hogan Lovells Cadwalader's organisational design is a template for how to scale agents without creating shadow systems or ungoverned workflows.
What this means for buyers and suppliers
For enterprise buyers, the Hogan Lovells Cadwalader announcement carries three practical takeaways.
First, agentic AI is now a core capability in professional services, not an experiment. When a top-ten global law firm goes firmwide with dual platforms and custom agent development, it sets a standard that procurement teams in other sectors will be expected to meet.
Second, the value of custom agents is in the workflow, not just the model. The competitive advantage comes from encoding your own expertise into reusable, governable agentic processes. Off-the-shelf agents can get you started. Proprietary agents are what separate leaders from followers.
Third, governance must keep pace. Building an AI board and commercial unit alongside the technology is recognition that the risk of agentic AI is not in the algorithms. It is in how decisions get made, how outputs get reviewed, and how accountability gets assigned when agents act on behalf of the organisation.
For suppliers, the story is equally clear. Buyers are not just asking for better models. They are asking for platforms that let them build, govern, and iterate their own agents. The supplier that helps a buyer turn institutional knowledge into agentic workflows will win long-term relationships. The supplier that treats AI as a chatbot upgrade will lose.
The Agentic Expo angle
Agentic Expo at Olympia London on 23 to 24 March 2027 is where the buyers and suppliers making these decisions come together. The Hogan Lovells Cadwalader story is exactly the kind of enterprise deployment that defines the market now: firmwide scale, dual-vendor strategy, custom agent development, and governance built in from the start.
If you are procuring agentic AI, this is what production looks like. If you are building it, this is the standard your buyers will measure you against.
Sources: Harvey, "Hogan Lovells Cadwalader Adopts Harvey", 16 October 2026; The Lawyer, "Hogan Lovells Cadwalader breaks ranks in new AI strategy with dual Legora and Harvey rollouts", 16 October 2026; Unite.AI, "Custom AI Agents Built by Hogan Lovells Cadwalader Lawyers, Harvey Says", 16 October 2026.